Condensed Milk Formulation Composition Calculation Verification Dataset

#Calculation verification #Numeric consistency judgment #Error type classification #Verification rationale generation #Condensed milk formulation review #Food quality control #Formulation calculation checks #Food product development

4600
0.3GB
JSON
CATL
MOBIUSI INC

Updated:2026-11-03

AI Analysis & Value Prop

This dataset focuses on composition calculations for condensed milk formulations, covering ingredient masses and compositions, final product mass, submitted numerical claims, and calculated metrics such as fat, sugar, and total solids. Each sample can also record the calculation basis, whether the claim is correct, the error type, and a verification rationale. It supports training and evaluating models that check formulation calculations, as well as food formulation review and calculation verification workflows.

Dataset Insights

Sample Examples

{}

Technical Specifications

Field	Type	Description
file_name	string	File name
error_type	string	Classifies the detected issue, or indicates that there is no error when the submitted claim is correct.
submitted_claim	object	Numerical composition values submitted for verification, optionally accompanied by an absolute tolerance in percentage points.
is_claim_correct	boolean	Indicates whether the submitted values agree with the calculated values within the specified tolerance.
calculation_basis	string	The formula or a concise explanation used to calculate final product composition, such as dividing component mass by final product mass and multiplying by 100.
formulation_input	object	Ingredient masses and composition data, together with the final product mass used in the calculation.
calculated_metrics	array	Composition percentages calculated from ingredient masses, ingredient composition, and final product mass.
verification_rationale	string	Explains the correctness decision and any identified discrepancy.

Compliance Statement

Authorization Type	Proprietary - Commercial AI Training License (No Redistribution)
Commercial Use	Requires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and Anonymization	No PII, no real company names, simulated scenarios follow industry standards
Compliance System	Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs

Frequently Asked Questions

What information does the dataset contain?: Samples can include ingredient masses and compositions for condensed milk formulations, final product mass, submitted numerical claims, calculated metrics, and calculation bases. They also record claim correctness, error types, and verification rationales. Metrics may include fat, sugar, or total solids.

What format is the data provided in?: The data is organized as JSON, with fields including formulation_input, submitted_claim, calculated_metrics, calculation_basis, is_claim_correct, error_type, and verification_rationale. This structure is suitable for processing and model training.

How can I access the dataset?: Check the product listing for its available download, access, or request options. The delivery format and availability are subject to the information on that listing.

What are the licensing and usage terms?: Refer to the license terms shown on the product listing for permitted use and restrictions. Confirm the applicable license before using the dataset in commercial projects or redistributing it.

Can't find the data you need?

Post a request and let data providers reach out to you.

Get this Dataset

Verified for Enterprise Use

Cite this Work

@dataset{Mobiusi2026,
  title={Condensed Milk Formulation Composition Calculation Verification Dataset},
  author={MOBIUSI INC},
  year={2026},
  url={https://www.mobiusi.com/datasets/74cb21d77f855a4e53d359016ac75d59},
  urldate={2026-10-26},
  keywords={condensed milk dataset,formulation composition calculation,recipe calculation verification,food calculation data,fat content calculation,sugar calculation,total solids calculation,outcome verifier training},
  version={1.0}
}

Using this in research? Please cite us.

placeholder

Scene

Domain

Modality

Task